From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 11 Mar 2004 15:25:52 -0800 From: Andrew Morton Subject: Re: blk_congestion_wait racy? Message-Id: <20040311152552.20a9bb06.akpm@osdl.org> In-Reply-To: References: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Martin Schwidefsky Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, piggin@cyberone.com.au List-ID: Martin Schwidefsky wrote: > > > An ouch-per-second sounds reasonable. It could simply be that the CPUs > > were off running other tasks - those timeout are less than scheduling > > quanta. > > I don't understand why an ouch-per-second is reasonable. The mempig is > the only process that runs on the machine and the blk_congestion_wait > uses HZ/10 as timeout value. I'd expect about 100 ouches for the 10 > seconds the test runs. blk_congestion_wait() is supposed to be terminated by someone releasing a disk write request. If no write requests are freed in 100 milliseconds then either Something Is Up or that process simply was not scheduled for some time after the wakeup was delivered. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: aart@kvack.org